Results 41 to 50 of about 129,018 (360)
Hierarchical Average Precision Training for Pertinent Image Retrieval [PDF]
ECCV 2022, Oct 2022, Tel-Aviv, Israel, 2022 Image Retrieval is commonly evaluated with Average Precision (AP) or Recall@k. Yet, those metrics, are limited to binary labels and do not take into account errors' severity. This paper introduces a new hierarchical AP training method for pertinent image retrieval (HAP-PIER). HAPPIER is based on a new H-AP metric, which leverages a concept hierarchy to
arxiv
Noisy Galvanic Stimulation Improves Roll-Tilt Vestibular Perception in Healthy Subjects
Frontiers in Neurology, 2018 It has recently been demonstrated that noisy galvanic vestibular stimulation (nGVS) delivered as imperceptible white noise can improve balance control via the induction of stochastic resonance.
Aram Keywan +4 more
doaj +1 more source
Robust and Decomposable Average Precision for Image Retrieval [PDF]
Thirty-fifth Conference on Neural Information Processing Systems
(NeurIPS 2021), Dec 2021, Sydney, Australia, 2021 In image retrieval, standard evaluation metrics rely on score ranking, e.g. average precision (AP). In this paper, we introduce a method for robust and decomposable average precision (ROADMAP) addressing two major challenges for end-to-end training of deep neural networks with AP: non-differentiability and non-decomposability. Firstly, we propose a new
arxiv
The “broken escalator” phenomenon: Vestibular dizziness interferes with locomotor adaptation [PDF]
, 2020 BACKGROUND: Although vestibular lesions degrade postural control we do not know the relative contributions of the magnitude of the vestibular loss and subjective vestibular symptoms to locomotor adaptation.
Arshad, Q. +13 more
core +2 more sources
Computer Animation and Virtual Worlds, Volume 33, Issue 6, November/December 2022., 2022 Stereo window violations in 3D rendering occur when virtual objects in front of a physical screen are clipped at the screen borders. Dynparity introduces a real‐time, per‐vertex, nonuniform projection to eliminate this adversary effect. Our user study shows that Dynparity significantly reduces simulator sickness compared to standard stereo rendering on
Christoph Schröder‐Dering +2 more
wiley +1 more source
Multi-Attribute Balanced Sampling for Disentangled GAN Controls [PDF]
arXiv, 2021 Various controls over the generated data can be extracted from the latent space of a pre-trained GAN, as it implicitly encodes the semantics of the training data. The discovered controls allow to vary semantic attributes in the generated images but usually lead to entangled edits that affect multiple attributes at the same time.
arxiv
Association Between Vestibular Migraine and Migraine Headache: Yet to Explore. [PDF]
, 2020 ObjectivesTo evaluate if patients with a diagnosis of vestibular migraine (VM) by the International Classification of Headache Disorders (ICHD) criteria have meaningful differences in symptomatology and disease characteristics when compared to patients ...
Abouzari, Mehdi +5 more
core
Frontiers in Aging Neuroscience, 2018 Objective: Many patients with dizziness and vertigo are of older age. It is still unclear which age-associated factors play a role in the treatment of dizziness and vertigo.
Maren Dietzek +8 more
doaj +1 more source
Scientific Reports, 2022 Neuronal lesions trigger mechanisms of structural and functional neuroplasticity, which can support recovery. However, the temporal and spatial appearance of structure–function changes and their interrelation remain unclear.
Melissa Antons +9 more
doaj +1 more source
Wasserstein Loss for Semantic Editing in the Latent Space of GANs [PDF]
arXiv, 2023 The latent space of GANs contains rich semantics reflecting the training data. Different methods propose to learn edits in latent space corresponding to semantic attributes, thus allowing to modify generated images. Most supervised methods rely on the guidance of classifiers to produce such edits.
arxiv